Clustering: which instance is in which cluster and extract these information

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Clustering: which instance is in which cluster and extract these information

Falko Guderian
Hello list members,

I'm working on my diploma thesis and want to cluster specific data from
different documents. I exported the data into arff format for example:
------------------------  file.arff  --------------------------
@relation weather

@attribute dokument {doc1, doc2, doc3, ...}
@attribute temperature real
@attribute humidity real
@attribute windy {TRUE, FALSE}
@attribute play {yes, no}
.
.
.

@data
doc1,85,85,FALSE,no,...
doc2,80,90,TRUE,no,...
doc3,95,50,FALSE,yes,...
....
------------------------  file.arff  --------------------------
Each instance stands for a specific document.

I used a clustering algorithm (for example EM) and could see the
coloured instances for each cluster in the visualization panel.
But I' m interested in the information which document (doc1, doc2,
doc3,...) is in which cluster.  I work with hunderts of documents and
I'd like to save the clustering information for further work.

I could extract these information for SimpleKMeans but I haven't found a
way for the other clustering algorithms.

I would be pleased if someone can help me.

Best regards

Falko Guderian


_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Discretize train-test individually

Rajni jain

Is it possible in Weka to discretize train first and use the split points
later to discretize the test set?
I tried searching the archive. Although similar Q was there but no solution?


Rajni Jain

_________________________________________________________________
Dhoom ya Humtum? http://server1.msn.co.in/sp05/iifa/ Which is your fav
movie?


_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Clustering: which instance is in which cluster and extract these information

Surendra Singhi
In reply to this post by Falko Guderian
Falko Guderian <[hidden email]> writes:

> Hello list members,
>
> I'm working on my diploma thesis and want to cluster specific data
> from different documents.

> But I' m interested in the information which document (doc1, doc2,
> doc3,...) is in which cluster.
> I'd like to save the clustering information for further work.
>
> I could extract these information for SimpleKMeans but I haven't found
> a way for the other clustering algorithms.
>
Look at the link it shows how to save this information.
http://www.public.asu.edu/~sksinghi/weka-faq-Z-H-5.html#node_sec_5.2

And if you want to use other clustering algorithms, replacing
SimpleKmeans by the other algorithms must be simple, just experiment a bit.  

HTH.

--
Surendra Singhi
http://www.public.asu.edu/~sksinghi/index.htm

Great wits are sure to madness near allied,
And thin partitions do their bounds divide.

   (John Dryden, Absalom and Achitophel, 1681)


_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
Loading...